Cacti / plugin_thold

Thold Plugin for Cacti
GNU General Public License v2.0
60 stars 60 forks source link

Thold 1.8.1 #675

Open botts99 opened 1 month ago

botts99 commented 1 month ago

Describe the bug A clear and concise description of what the bug is. It appears running thold 1.8.1 on cacti 1.2.26 with remote pollers causes an issue where one a large amount of devices go offline, it never recovers and sends out emails about downed hosts but never stop. Disabling the plugin stops the emails.

Screenshots If applicable, add screenshots to help explain your problem.

Plugin (please complete the following information):

Most devices recover but 4 do not. image

Thold continues to crash and time out so alerting continues untill disabled. image

Login and disable thold and recovers image

I did just notice though that thold comes back onine for some reason so I had to disable it a second time.

Prior to those events it runs and processes fine.

image

bmfmancini commented 1 month ago

So are you seeing the same email being sent over and over again or a down and restore?

What version of. Php and what os?

On Sat, Jun 8, 2024, 10:56 botts99 @.***> wrote:

Describe the bug A clear and concise description of what the bug is. It appears running thold 1.8.1 on cacti 1.2.26 with remote pollers causes an issue where one a large amount of devices go offline, it never recovers and sends out emails about downed hosts but never stop. Disabling the plugin stops the emails.

Screenshots If applicable, add screenshots to help explain your problem.

Plugin (please complete the following information):

Most devices recover but 4 do not. image.png (view on web) https://github.com/Cacti/plugin_thold/assets/59628262/72953078-a758-448e-8288-74e149a07955

Thold continues to crash and time out so alerting continues untill disabled. image.png (view on web) https://github.com/Cacti/plugin_thold/assets/59628262/00e9fbde-37fd-4813-ad2a-65c7426f32fc

Login and disable thold and recovers image.png (view on web) https://github.com/Cacti/plugin_thold/assets/59628262/299263e8-747f-4b70-b328-fc353e3f60f0

I did just notice though that thold comes back onine for some reason so I had to disable it a second time.

— Reply to this email directly, view it on GitHub https://github.com/Cacti/plugin_thold/issues/675, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADGEXTELKJKRKMJZ7JXXCF3ZGMLRZAVCNFSM6AAAAABJABRXLKVHI2DSMVQWIX3LMV43ASLTON2WKOZSGM2DCNRZGIZDEMI . You are receiving this because you are subscribed to this thread.Message ID: @.***>

botts99 commented 1 month ago

Looks to be the restore email. Will get the php version shortly.

PHP 7.0.33-0+deb9u12

On Sat, Jun 8, 2024, 11:05 AM Sean Mancini @.***> wrote:

So are you seeing the same email being sent over and over again or a down and restore?

What version of. Php and what os?

On Sat, Jun 8, 2024, 10:56 botts99 @.***> wrote:

Describe the bug A clear and concise description of what the bug is. It appears running thold 1.8.1 on cacti 1.2.26 with remote pollers causes an issue where one a large amount of devices go offline, it never recovers and sends out emails about downed hosts but never stop. Disabling the plugin stops the emails.

Screenshots If applicable, add screenshots to help explain your problem.

Plugin (please complete the following information):

Most devices recover but 4 do not. image.png (view on web) < https://github.com/Cacti/plugin_thold/assets/59628262/72953078-a758-448e-8288-74e149a07955>

Thold continues to crash and time out so alerting continues untill disabled. image.png (view on web) < https://github.com/Cacti/plugin_thold/assets/59628262/00e9fbde-37fd-4813-ad2a-65c7426f32fc>

Login and disable thold and recovers image.png (view on web) < https://github.com/Cacti/plugin_thold/assets/59628262/299263e8-747f-4b70-b328-fc353e3f60f0>

I did just notice though that thold comes back onine for some reason so I had to disable it a second time.

— Reply to this email directly, view it on GitHub https://github.com/Cacti/plugin_thold/issues/675, or unsubscribe < https://github.com/notifications/unsubscribe-auth/ADGEXTELKJKRKMJZ7JXXCF3ZGMLRZAVCNFSM6AAAAABJABRXLKVHI2DSMVQWIX3LMV43ASLTON2WKOZSGM2DCNRZGIZDEMI>

. You are receiving this because you are subscribed to this thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/Cacti/plugin_thold/issues/675#issuecomment-2156087939, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOG5VZQ6HIJHQIAY4PSMPT3ZGMTUNAVCNFSM6AAAAABJABRXLKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNJWGA4DOOJTHE . You are receiving this because you authored the thread.Message ID: @.***>

botts99 commented 1 month ago

Here is more of the log strange that it cant pull data but if I go to those devices they are online.

Seems to do that over and over till I disable thold and turn back on.

One a side note, I updated thold from 1.6.0 to the new version just a few days ago. Had no issues on 1.6.0 but assume there have been many changes since then.

image

botts99 commented 1 month ago

Logs from the remote poller

Capture

TheWitness commented 2 weeks ago

@botts99,

What is the availability method that you are using for these devices?

TheWitness commented 2 weeks ago

Also, what are those snmp2_get() calls from? Some script I imagine. Are you using the Notification Queue in the latest THOLD also, have you selected to receive a single Email notification?

image

TheWitness commented 2 weeks ago

These is also a feature that was withdrawn due to timing about suspending notification when X devices at a site go down. Next release I guess.

botts99 commented 2 weeks ago

I upgraded from 1.60 to the current version so perhaps somthing didnt update correctly. I remove everything with thold and resetup the the currnet release and have not had an issue since. I am wondeing if somthing didnt populate right to the remote poller and since I did a complete remove and setup if that corrected those issue.

image

TheWitness commented 2 weeks ago

Well, it's not working, but it will soon. Look for a commit. I'll reference it here.

TheWitness commented 2 weeks ago

Okay, you should have better luck now. Some of your Emails may be coming from Monitor. It's something you can disable too.